Low daily step counts tied to dementia, Parkinson's and stroke in Fitbit study

What happened

People who averaged fewer than 5,000 steps a day had 80% higher rates of dementia, Parkinson's disease and stroke than people averaging more than 12,000. The work is due to be presented on 20 October 2026 in San Diego at an American Neurological Association meeting, as first reported by Medical Xpress (Medical Xpress).

A caveat up front: we could not locate the abstract or a press release. Everything below about the new study comes from that news item, which is a HealthDay story. Treat the numbers as secondhand until the researchers publish details.

What the study reportedly found

The researchers analyzed a year of Fitbit data from more than 22,000 adults aged 40 and older. They then checked those people's medical records for another five years to see who developed neurological disease.

• Fewer than 5,000 steps a day: rates of dementia, Parkinson's and stroke were 80% higher than in the group averaging more than 12,000.
• About 8,000 steps a day: lower rates of neurological disease than the least-active group.
• Least-active group: also had more obesity, diabetes, high blood pressure and depression, which the researchers say may explain some of the gap.

The lead author is Dr. Meet Popatbhai Kachhadia, a neurology resident at Florida Atlantic University. Co-author Dr. Marc Swerdloff is at FAU and the Marcus Neuroscience Institute.

Swerdloff called 8,000 steps a realistic target and said it lines up with federal guidance of roughly 150 minutes of moderate activity a week. He also said nobody needs to hit 12,000. That comparison is his, not a result from the study.

What it does not show

Start with the obvious: this is observational. It can show that low step counts and brain disease travel together. It cannot show that walking more prevents them.

Reverse causation is the big worry. Early, undiagnosed neurological disease can make people move less, so low steps may be an early symptom rather than a cause. The news item does not say whether the researchers accounted for that, for example by excluding early events.

Confounding is the second problem. The least-active group had more obesity, diabetes, high blood pressure and depression, all of which are tied to brain health on their own. The researchers say these may explain part of the difference.

Finally, research presented at meetings is considered preliminary until it is published. The researchers themselves say more studies are needed to confirm the link.

What it means for you

Your phone or watch already tracks steps, so checking your weekly average costs nothing. That is a reasonable first move. It is not a brain-health test.

The researchers' own framing is cautious. Consistently low step counts should prompt an overall health check, not fear about your brain. If your average sits well under 5,000 and you don't know why, that is worth raising with a clinician, along with blood pressure, blood sugar, weight and mood.

If you want a target, the reported one is about 8,000 steps a day, which was associated with lower rates than the least-active group. You do not need 12,000. Nothing here shows that hitting 8,000 would change your own risk, so think of it as a sensible activity goal rather than a protective dose.

Context: other step-count research

Wearable-based step research is not new. A 2022 Nature Medicine paper used All of Us Research Program data linked to health records, with 6,042 participants who had their own Fitbit data. Their median was 7,731 steps a day over a median of 4.0 years of monitoring. That sample was 73% female, 84% white and 71% college-educated.

We can't confirm the new study uses All of Us. If it does, that skew would limit how far the results generalize, so it is a question for the authors.

A JAMA Neurology paper on daily step count and incident dementia in 78,430 UK adults also exists. We have seen only its title and have not reviewed its findings, so we are not leaning on it here.